Emerging evidence demonstrates that Autism Spectrum Disorder (ASD) can be reliably diagnosed by age two, and that early identification and intervention can improve outcomes. Low-income and minority children with ASD, however, are diagnosed later and experience greater delays in service provision than their white and more financially advantaged peers. Feasible, culturally appropriate interventions with broad scale-up potential are necessary to reduce this disparity. This R01 application builds upon pilot studies (Augustyn, HRSA R40 MC19928; Feinberg, AHRQ R03 HS22155) of an adapted version of Patient Navigation as means to reduce disparities in ASD diagnosis and service provision. Patient Navigation is a lay-delivered case management approach that focuses on overcoming logistical hurdles to care during a defined episode. Pilot data from our research group demonstrate the feasibility of an adapted approach of Patient Navigation, referred to as Family Navigation (FN), among families of children with suspected ASD, and provide suggestive evidence that the approach both reduces time to ASD diagnosis and increases engagement with services. We propose a multisite, randomized comparative effectiveness trial (n=19,500) of a systemic, lay-delivered FN protocol, which begins prior to a child's 18 or 24 month health supervision visit and ends 100 days after an ASD diagnosis is made. The trial takes place in urban, integrated care networks that provide healthcare to low-income children. The basic structure of both intervention arms is a collaborative care system. The conventional care management arm (CCM) is consistent with the type of care provided within a traditional - but high quality - medical home. The FN arm provides more intensive, individually tailored, care coordination and theory-based family support. Children will be followed for 12 months. Data regarding screening outcomes, diagnosis and service utilization will be abstracted from children's medical records. Measures of parental stress, self-management skills, caregiver burden, and satisfaction with services will be administered over 4 collection time points, linked to key intervention outcomes. We will assess the superiority of FN as compared to CCM as a means to: achieve an 80% screening rate for indicators of ASD across all sites; implement a decision rule for referral to ASD evaluation; shorten the time to diagnosis among children suspected to have ASD; shorten the time to deployment of ASD services among those diagnosed; and improve engagement with ASD services. Our lay-delivered FN system has been designed to have broad scale-up potential. Thus, to provide critical information on how such a system could be best implemented and disseminated across urban primary care settings nationwide, we will conduct a concurrent implementation analysis - systematically examining performance gaps in FN's delivery according to the Theoretical Domains Framework. If successful, our study will provide real world primary care practices with a replicable model of care that increases early identification and access to timely diagnostic and early intervention services for a vulnerable population of urban families.